Improve Autocomplete Performance on my package


#1

Hello Atom Package Developers,

I developed the https://atom.io/packages/wordpress-api package.

There are few issues users are reporting around slow performance. I have been looking for ways to fix that but I have not found an approach to that I am happy with to address the issues.

I would really appreciate if anyone can provide some ideas or tell me what I have been doing wrong.

Kind Regards,

Azizur


#2

Is it all snippets? It is hard to imagine how they could be slow. I’d try it but I don’t know wordpress.

Are you using the built-in snippet feature or watching the text yourself? Watching it with your own code could very well be slow.


#3

Thanks @mark_hahn for your reply.

Only thing I can think of is the size of the one of auto complete grammar file: wordpress.functions.cson. I thought it would be loaded on startup into memory for reuse during type hinting.

I wonder if I can split that file into smaller chunk or perhaps there is other way to optimise this file for performance.


#4

I looked in the repo and your files are tiny. They can’t be the problem. Are your users reporting slowdown at startup or while they are using it later? What other package does your package rely on? There is more than one way to do autocomplete on Atom.

Unfortunately I can only help so much here because I’ve never worked on a package like this. Maybe if you describe the slowness and situation in as much detail as you can someone here that knows more can help.


#5

Thanks @mark_hahn,

I have personally not seen the slowness. However few users are reporting issue when they are using the package.

A list of issues can be seen at https://github.com/azizur/atom-wordpress-api/issues